Simulation of log fitness change
Laure Olazcuaga, Nicolas Rode
02-December-2020
1 DATA
## Longitudinal data
data_sum <- loadlongitudinaldata(dataset = "DATA_Adults_G1G29.csv", rm_generation1 = 1,rm_generation2 = 7,rm_generation3 = 29)
## Phenotyping steps
data_G0 <- loadfitnessdata(dataset = "Selection_Phenotypage_G0_G7_G8.csv", generation = "G1")
data_G7 <- loadfitnessdata(dataset = "Selection_Phenotypage_G0_G7_G8.csv", generation = "G7")
data_G29 <- loadfitnessdata(dataset =
"PERFORMANCE_Comptage_adultes_G13G14G15G16G17G18G19G20G21G22G23G24G25G26G27G28G29.csv",
generation = "29")2 SUP: Pairwise of Fitness difference
2.1 Data
#Compute log change
data_logchange<-computelogchange(fitness_dataset_intermediate = data_G7, fitness_dataset_final = data_G29)
#Formatting for plotting pairwise
EC_Cran_Cher_G7<-formattingplotpair(data_logchange, "7", "Cherry", "Cranberry")
EC_Straw_Cran_G7<-formattingplotpair(data_logchange, "7", "Cranberry", "Strawberry")
EC_Straw_Cher_G7<-formattingplotpair(data_logchange, "7", "Strawberry", "Cherry")
EC_Cran_Cher_G29<-formattingplotpair(data_logchange, "29", "Cherry", "Cranberry")
EC_Straw_Cran_G29<-formattingplotpair(data_logchange, "29", "Cranberry", "Strawberry")
EC_Straw_Cher_G29<-formattingplotpair(data_logchange, "29", "Strawberry", "Cherry")2.2 Plot
#Limit axis
lim_G7<-max(abs(min(data_logchange[data_logchange$Generation=="7",]$logchange-
1.96*data_logchange[data_logchange$Generation=="7",]$sd_logchange, na.rm = T)),
max(data_logchange[data_logchange$Generation=="7",]$logchange+
1.96*data_logchange[data_logchange$Generation=="7",]$sd_logchange, na.rm = T))
lim_G29<-max(abs(min(data_logchange[data_logchange$Generation=="29",]$logchange-
1.96*data_logchange[data_logchange$Generation=="29",]$sd_logchange, na.rm = T)),
max(data_logchange[data_logchange$Generation=="29",]$logchange+
1.96*data_logchange[data_logchange$Generation=="29",]$sd_logchange, na.rm = T))
#Plot
PAIR_StrawCran_G7<-ggplot(data = EC_Straw_Cran_G7,
aes(x = logchange_Strawberry,y = logchange_Cranberry, color=Fruit_s, group = Line)) +
geom_errorbar(aes(ymin = logchange_Cranberry-1.96*sd_logchange_Cranberry,
ymax = logchange_Cranberry+1.96*sd_logchange_Cranberry),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(xmin = logchange_Strawberry-1.96*sd_logchange_Strawberry,
xmax = logchange_Strawberry+1.96*sd_logchange_Strawberry),
height=0.02,size=0.2,alpha=1) +
geom_point(shape=21, size=3, fill = "#ffffff", stroke=1.3) +
xlab("Fitness difference\nin strawberry") +
ylab("Fitness difference\nin cranberry") +
xlim(-lim_G7,lim_G7) +
ylim(-lim_G7,lim_G7) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
theme(axis.title.x = element_text(size=10, colour = "#3FAA96"),
axis.title.y = element_text(size=10, colour = "#FDB424")) +
theme_LO_sober
PAIR_StrawCran_G7
PAIR_StrawChe_G7<-ggplot(data = EC_Straw_Cher_G7,
aes(x = logchange_Cherry,y = logchange_Strawberry, group=Line, color=Fruit_s)) +
geom_errorbar(aes(ymin = logchange_Strawberry-1.96*sd_logchange_Strawberry,
ymax = logchange_Strawberry+1.96*sd_logchange_Strawberry),
width=0.02, size=0.2, alpha=1) +
geom_errorbarh(aes(xmin = logchange_Cherry-1.96*sd_logchange_Cherry,
xmax = logchange_Cherry+1.96*sd_logchange_Cherry),
height=0.02, size=0.2, alpha=1) +
geom_point(shape=21, size=3, fill = "#ffffff", stroke=1.3) +
ylab("Fitness difference\nin strawberry") +
xlab("Fitness difference\nin cherry") +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
xlim(-lim_G7,lim_G7) +
ylim(-lim_G7,lim_G7) +
theme(axis.title.x = element_text(size=10, colour = "#BC3C6D"),
axis.title.y = element_text(size=10, colour = "#3FAA96")) +
theme_LO_sober
PAIR_StrawChe_G7
PAIR_CheCran_G7<-ggplot(data = EC_Cran_Cher_G7,aes(x = logchange_Cranberry,y = logchange_Cherry,
group=Line, color=Fruit_s)) +
geom_errorbar(aes(ymin = logchange_Cherry-1.96*sd_logchange_Cherry,
ymax = logchange_Cherry+1.96*sd_logchange_Cherry),
width=0.02, size=0.2, alpha=1) +
geom_errorbarh(aes(xmin = logchange_Cranberry-1.96*sd_logchange_Cranberry,
xmax = logchange_Cranberry+1.96*sd_logchange_Cranberry),
height=0.02, size=0.2, alpha=1) +
geom_point(shape=21, size=3, fill = "#ffffff", stroke=1.3) +
ylab("Fitness difference\nin cherry") +
xlab("Fitness difference\nin cranberry") +
xlim(-lim_G7, lim_G7) +
ylim(-lim_G7, lim_G7) +
scale_color_manual(values = c("#BC3C6D", "#FDB424", "#3FAA96")) +
theme_LO_sober +
theme(axis.title.x = element_text(colour = "#FDB424"),
axis.title.y = element_text(colour = "#BC3C6D"))
PAIR_CheCran_G7
################## Final phenotyping step
#Plot
PAIR_CranStraw_G29<-ggplot(data = EC_Straw_Cran_G29,
aes(x = logchange_Strawberry, y = logchange_Cranberry, color=Fruit_s,group=Line)) +
geom_errorbar(aes(ymin = logchange_Cranberry-1.96*sd_logchange_Cranberry,
ymax = logchange_Cranberry+1.96*sd_logchange_Cranberry),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(xmin = logchange_Strawberry-1.96*sd_logchange_Strawberry,
xmax = logchange_Strawberry+1.96*sd_logchange_Strawberry),
height=0.02,size=0.2,alpha=1) +
geom_point(aes(fill = Fruit_s),shape=21, size=3) +
xlab("Fitness difference\nin strawberry") +
ylab("Fitness difference\nin cranberry") +
xlim(-lim_G29,lim_G29) +
ylim(-lim_G29,lim_G29) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_fill_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
theme(axis.title.x = element_text(size=10, colour = "#3FAA96"),
axis.title.y = element_text(size=10, colour = "#FDB424")) +
theme_LO_sober
PAIR_CranStraw_G29
PAIR_StrawChe_G29<-ggplot(data = EC_Straw_Cher_G29,
aes(x = logchange_Cherry,y = logchange_Strawberry, group=Line , color=Fruit_s)) +
geom_errorbar(aes(ymin = logchange_Strawberry-1.96*sd_logchange_Strawberry,
ymax = logchange_Strawberry+1.96*sd_logchange_Strawberry),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(xmin = logchange_Cherry-1.96*sd_logchange_Cherry,
xmax = logchange_Cherry+1.96*sd_logchange_Cherry),
height=0.02,size=0.2,alpha=1) +
geom_point(aes(fill = Fruit_s),shape=21, size=3) +
ylab("Fitness difference\nin strawberry") +
xlab("Fitness difference\nin cherry") +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_fill_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
xlim(-lim_G29,lim_G29) +
ylim(-lim_G29,lim_G29) +
theme(axis.title.x = element_text(colour = "#BC3C6D"),
axis.title.y = element_text(colour = "#3FAA96")) +
theme_LO_sober
PAIR_StrawChe_G29
PAIR_CheCran_G29<-ggplot(data = EC_Cran_Cher_G29,
aes(x = logchange_Cranberry,y = logchange_Cherry,group=Line, color=Fruit_s)) +
geom_errorbar(aes(ymin = logchange_Cherry-1.96*sd_logchange_Cherry,
ymax = logchange_Cherry+1.96*sd_logchange_Cherry),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(xmin = logchange_Cranberry-1.96*sd_logchange_Cranberry,
xmax = logchange_Cranberry+1.96*sd_logchange_Cranberry),
height=0.02,size=0.2,alpha=1) +
geom_point(aes(fill = Fruit_s), shape=21, size=3) +
ylab("Fitness difference\nin cherry") +
xlab("Fitness difference\nin cranberry") +
xlim(-lim_G29,lim_G29) +
ylim(-lim_G29,lim_G29) +
scale_fill_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
theme_LO_sober +
theme(axis.title.x = element_text(colour = "#FDB424"),
axis.title.y = element_text(colour = "#BC3C6D"))
PAIR_CheCran_G29
#Legend
TEMP_data_logchange_sum_Straw_Cher_G7<-data_logchange
TEMP_data_logchange_sum_Straw_Cher_G7$Pheno<-sample(c("Intermediate", "Final"),
length(data_logchange$logchange),
replace=TRUE)
PAIR_plot_legend<- ggplot(data=TEMP_data_logchange_sum_Straw_Cher_G7,
aes(x=Treatment, y=logchange, group=Fruit_s, colour=Fruit_s,shape=Pheno)) +
geom_errorbar(aes(ymin=logchange-1.96*sd_logchange, ymax=logchange+1.96*sd_logchange),
width=0.01,size=1) +
geom_point(size=3, stroke=1.3) +
scale_shape_manual(name = "Phenotyping step",
labels = c("Intermediate", "Final"),values=c(21,16)) +
scale_color_manual(name= "Selection fruit",
values=c("#BC3C6D", "#FDB424", "#3FAA96"),
label=c("Cherry", "Cranberry", "Strawberry")) +
theme_LO_sober
PAIR_legend<-lemon::g_legend(PAIR_plot_legend)
rm(TEMP_data_logchange_sum_Straw_Cher_G7)
PAIR_Fig_Sup <- cowplot::ggdraw() +
cowplot::draw_plot(PAIR_CheCran_G7+theme(legend.position = "none"),
x =0.00, y = 0.5, width = 0.22, height =0.45) +
cowplot::draw_plot(PAIR_StrawCran_G7+theme(legend.position = "none"),
x = 0.26, y = 0.5, width = 0.22, height = 0.45) +
cowplot::draw_plot(PAIR_StrawChe_G7+theme(legend.position = "none"),
x = 0.52, y = 0.5, width = 0.22, height = 0.45) +
cowplot::draw_plot(PAIR_legend, x = 0.80, y = 0.5, width = 0.1, height = 0.1) +
cowplot::draw_plot(PAIR_CheCran_G29+theme(legend.position = "none"),
x =0.00, y = 0, width = 0.22, height = 0.45) +
cowplot::draw_plot(PAIR_CranStraw_G29+theme(legend.position = "none"),
x = 0.26, y = 0, width = 0.22, height = 0.45) +
cowplot::draw_plot(PAIR_StrawChe_G29+theme(legend.position = "none"),
x = 0.52, y = 0, width = 0.22, height = 0.45) +
cowplot::draw_plot_label(c("Intermediate phenotyping step", "A", "B", "C", " ",
"Final phenotyping step", "D", "E", "F", " "),
x = c(0.26,0,0.26,0.52,0.82,0.2,0,0.26,0.52,0.82),
y = c(1.01,0.97,0.97,0.97,0.97,0.5,0.47, 0.47, 0.47, 0.47),
hjust = c(-0.25,-0.25,-0.25,-0.25,-0.25,-0.75,-0.75,-0.75,-0.75,-0.75),
vjust = c(1.5,1.5,1.5,1.5,1.5,1.5,1.5,1.5,1.5,1.5),
size = 14)
PAIR_Fig_Sup
cowplot::save_plot(file=here::here("figures", "SUPMAT_LogChange_Pairwise.pdf"), PAIR_Fig_Sup,
base_height = 15/cm(1), base_width = 30/cm(1), dpi = 610)3 SUP: Long vs Fitness
3.1 Data
#To consider fruit maternal effect and GF maternal effect: new plot
## Y-axis = log change = log(Fitness pheno)-log(fitness initial)
## X-axis = fitness temp / fitness initial
## example: for CEA during first phase
## log(mean (CE1 from G2 to G5) / mean (Cherry G2))
##To check with Nico: for x-axis: denominator=mean Cherry G2 or mean CE1 G2 (???????????)
####### Phenotyping steps
#Compute log change
data_logchange <- computelogchange(fitness_dataset_intermediate = data_G7, fitness_dataset_final = data_G29)
#Remove allopatric data
data_fitness_pheno <- data_logchange[data_logchange$SA==1,]
data_fitness_pheno <- data_fitness_pheno[, c("Line", "Fruit_s", "logchange", "sd_logchange")]
####### Longitudinal data
#Compute log change
data_logchange_long <- computelogchange_forlongdata(longitudinal_dataset = data_sum)
####### Merge data
data_compare_longitudinal_pheno <- merge(data_logchange_long,data_fitness_pheno,by=c("Line", "Fruit_s"))
data_compare_longitudinal_pheno <- droplevels(data_compare_longitudinal_pheno)
## Does not seem to be a correlation between measures
cor.test(data_compare_longitudinal_pheno$logchange[data_compare_longitudinal_pheno$Phase=="first_prepool"], data_compare_longitudinal_pheno$logchange_long[data_compare_longitudinal_pheno$Phase=="first_prepool"])##
## Pearson's product-moment correlation
##
## data: data_compare_longitudinal_pheno$logchange[data_compare_longitudinal_pheno$Phase == "first_prepool"] and data_compare_longitudinal_pheno$logchange_long[data_compare_longitudinal_pheno$Phase == "first_prepool"]
## t = -0.37623, df = 12, p-value = 0.7133
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.6039532 0.4482880
## sample estimates:
## cor
## -0.1079732cor.test(data_compare_longitudinal_pheno$logchange[data_compare_longitudinal_pheno$Phase=="second_postpool"], data_compare_longitudinal_pheno$logchange_long[data_compare_longitudinal_pheno$Phase=="second_postpool"])##
## Pearson's product-moment correlation
##
## data: data_compare_longitudinal_pheno$logchange[data_compare_longitudinal_pheno$Phase == "second_postpool"] and data_compare_longitudinal_pheno$logchange_long[data_compare_longitudinal_pheno$Phase == "second_postpool"]
## t = 0.91427, df = 9, p-value = 0.3844
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.3737080 0.7587141
## sample estimates:
## cor
## 0.2915193####### Geary test
#Lines: CE2 and CR2 do not pass Geary's test (the threshold of a standardized mean greater than 3)
data_compare_longitudinal_pheno$Line_type <- ifelse(data_compare_longitudinal_pheno$Line == "CR2"|
data_compare_longitudinal_pheno$Line == "CE2", "solid","dashed")3.2 PLOT Step=f(Phase)
PLOT_IntermediatePheno_LOGCHANGE<-ggplot(data=data_compare_longitudinal_pheno[data_compare_longitudinal_pheno$Phase=="first_prepool",],
aes(y=logchange,x=logchange_long, color=Fruit_s,shape=Fruit_s, linetype=Line_type)) +
geom_errorbar(aes(x = logchange_long,
ymin = logchange-(1.96*sd_logchange),
ymax = logchange+(1.96*sd_logchange)),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(y = logchange,
xmin = logchange_long-1.96*sd_logchange_long,
xmax = logchange_long+1.96*sd_logchange_long),
height=0.02,size=0.2,alpha=1) +
geom_point(size=3, fill = "#ffffff", stroke=1.3) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_shape_manual(values=c(21, 22, 24)) +
xlab("Fitness difference between average fitness\nacross generations 2 to 5 and generation 2") +
ylab("Fitness difference between\nintermediate and initial phenotyping steps") +
labs(color="Selection fruit", shape="Selection fruit") +
geom_abline(intercept = 0, slope = 1, color="black", linetype="dashed", size=0.8) +
ylim(-1.9,0.9) +
xlim(-1.9,0.9) +
guides(linetype=FALSE) +
ggtitle("Intermediate phenotyping step vs. Phase 1") +
theme_LO_sober
PLOT_IntermediatePheno_LOGCHANGE 
PLOT_FinalPheno_LOGCHANGE <-
ggplot(data=data_compare_longitudinal_pheno
[data_compare_longitudinal_pheno$Phase=="second_postpool",],
aes(y=logchange, x=logchange_long,
color=Fruit_s, shape=Fruit_s, fill=Fruit_s, linetype=Line_type)) +
geom_errorbar(aes(x = logchange_long,
ymin = logchange-(1.96*sd_logchange),
ymax = logchange+(1.96*sd_logchange)),
width=0.02,size=0.2,alpha=1) +
geom_errorbarh(aes(y = logchange,
xmin = logchange_long-1.96*sd_logchange_long,
xmax = logchange_long+1.96*sd_logchange_long),
height=0.02,size=0.2,alpha=1) +
geom_abline(intercept = 0, slope = 1, color="black",
linetype="dashed", size=0.8) +
geom_point(size=3, stroke=1) +
scale_fill_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_shape_manual(values=c(21, 22, 24)) +
xlab("Fitness difference between average fitness\nacross generations 13 to 26 and generation 2") +
ylab("Fitness difference between\nfinal and initial phenotyping steps") +
labs(color="Selection fruit", shape="Selection fruit", fill="Selection fruit") +
ylim(-0.5,1.7) +
xlim(-0.5,1.7) +
guides(linetype=FALSE) +
ggtitle("Final phenotyping step vs. Phase 3") +
theme_LO_sober
PLOT_FinalPheno_LOGCHANGE 
PLOT_Comparison_Pheno_Longitudinal_LOGCHANGE <- cowplot::ggdraw() +
cowplot::draw_plot(PLOT_IntermediatePheno_LOGCHANGE +
theme(axis.title.y = element_text(size=10),
axis.title.x = element_text(size=10)),
x = 0, y = 0.5, width = 1, height = 0.5) +
cowplot::draw_plot(PLOT_FinalPheno_LOGCHANGE +
theme(axis.title.y = element_text(size=10),
axis.title.x = element_text(size=10)),
x =0, y = 0, width = 1, height = 0.5) +
cowplot::draw_plot_label(c("A", "B"),
x = c(0,0), y = c(1,0.5),
size = 14)
PLOT_Comparison_Pheno_Longitudinal_LOGCHANGE
cowplot::save_plot(file=here::here("figures", "SUPMAT_LogChange_Pheno_Longitudinal.pdf"),
PLOT_Comparison_Pheno_Longitudinal_LOGCHANGE, base_height = 20/cm(1), base_width = 15/cm(1), dpi = 1200)3.3 PLOT Step/Phase=f(Phase)
PLOT_IntermediatePheno_TEMP<-ggplot(data=data_compare_longitudinal_pheno[data_compare_longitudinal_pheno$Phase=="first_prepool",],
aes(y=logchange/logchange_long,x=logchange_long, color=Fruit_s,shape=Fruit_s)) +
geom_errorbarh(aes(y = logchange/logchange_long,
xmin = logchange_long-1.96*sd_logchange_long,
xmax = logchange_long+1.96*sd_logchange_long,
linetype=Line_type),
height=0.02,size=0.2,alpha=1) +
geom_point(size=3, fill = "#ffffff", stroke=1.3) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_shape_manual(values=c(21, 22, 24)) +
xlab("Fitness difference between average fitness\nacross generations 2 to 5 and generation 2") +
ylab("Fitness difference between phenotyping steps /\nFitness difference between phases") +
labs(color="Selection fruit", shape="Selection fruit") +
geom_abline(intercept = 0, slope = 1, color="black", linetype="dashed", size=0.8) +
ylim(-3.9,4.4) +
xlim(-3.9,4.4) +
ggtitle("Intermediate phenotyping step vs. Phase 1") +
theme_LO_sober
PLOT_IntermediatePheno_TEMP 
PLOT_FinalPheno_TEMP <- ggplot(data=data_compare_longitudinal_pheno
[data_compare_longitudinal_pheno$Phase=="second_postpool",],
aes(y=logchange/logchange_long,x=logchange_long, color=Fruit_s,shape=Fruit_s,fill=Fruit_s)) +
geom_errorbarh(aes(y = logchange/logchange_long,
xmin = logchange_long-1.96*sd_logchange_long,
xmax = logchange_long+1.96*sd_logchange_long,
linetype=Line_type),
height=0.02,size=0.2,alpha=1) +
geom_abline(intercept = 0, slope = 1, color="black",
linetype="dashed", size=0.8) +
geom_point(size=3, stroke=1) +
scale_fill_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_shape_manual(values=c(21, 22, 24)) +
xlab("Fitness difference between average fitness\nacross generations 13 to 26 and generation 2") +
ylab("Fitness difference between phenotyping steps /\nFitness difference between phases") +
labs(color="Selection fruit", shape="Selection fruit", fill="Selection fruit") +
ylim(-0.5,1.7) +
xlim(-0.5,1.7) +
ggtitle("Final phenotyping step vs. Phase 3") +
theme_LO_sober
PLOT_FinalPheno_TEMP 
PLOT_Comparison_TEMP <- cowplot::ggdraw() +
cowplot::draw_plot(PLOT_IntermediatePheno_TEMP +
theme(axis.title.y = element_text(size=10),
axis.title.x = element_text(size=10)),
x = 0, y = 0.5, width = 1, height = 0.5) +
cowplot::draw_plot(PLOT_FinalPheno_TEMP +
theme(axis.title.y = element_text(size=10),
axis.title.x = element_text(size=10)),
x =0, y = 0, width = 1, height = 0.5) +
cowplot::draw_plot_label(c("A", "B"),
x = c(0,0), y = c(1,0.5),
size = 14)
PLOT_Comparison_TEMP
cowplot::save_plot(file=here::here("figures", "SUPMAT_LogChange_Pheno-div-Long_Longitudinal.pdf"),
PLOT_Comparison_TEMP, base_height = 20/cm(1), base_width = 15/cm(1), dpi = 1200)3.4 PLOT Maternal effects
m <- lm(logchange ~ logchange_long + offset(logchange_long),
data = data_compare_longitudinal_pheno[data_compare_longitudinal_pheno$Phase=="second_postpool",])
# le summary permet directement de vérifier si la pente est significativement différente de 1 à partir de la p-value associée à la variable logchange_long et de tester si l'évolution du changement maternel est significatif (p-value associée à l'intercept)
summary(m)##
## Call:
## lm(formula = logchange ~ logchange_long + offset(logchange_long),
## data = data_compare_longitudinal_pheno[data_compare_longitudinal_pheno$Phase ==
## "second_postpool", ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.31398 -0.17353 -0.05823 0.12347 0.53029
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.05477 0.21373 0.256 0.80351
## logchange_long -0.81441 0.20300 -4.012 0.00306 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2743 on 9 degrees of freedom
## Multiple R-squared: 0.08498, Adjusted R-squared: -0.01669
## F-statistic: 0.8359 on 1 and 9 DF, p-value: 0.3844#pval of logchange_long = 0.003 => slope is different of 1
#pval of Intercept = 0.80 => no evolution of maternal effect???
## Ratio effet maternels pop dérivée sur effet maternels pop ancestrale??
exp(coef(m)[1]) ## (Intercept)
## 1.056301# proportion of adaptation due to maternal effects
data_compare_longitudinal_pheno$mateff <- (data_compare_longitudinal_pheno$logchange_long -
data_compare_longitudinal_pheno$logchange) / data_compare_longitudinal_pheno$logchange_long
data_compare_longitudinal_pheno## Line Fruit_s Phase logchange_long sd_logchange_long logchange
## 1 CE1 Cherry first_prepool 0.33816796 0.04332430 -0.27403677
## 2 CE2 Cherry first_prepool 0.14436267 0.05963021 -0.56171884
## 3 CE3 Cherry first_prepool 0.46747518 0.05873193 0.18528246
## 4 CE4 Cherry first_prepool 0.39817852 0.03355308 0.12791340
## 5 CEA Cherry second_postpool 1.57245790 0.02955303 0.14767000
## 6 CEB Cherry second_postpool 1.55711969 0.02872255 0.25038029
## 7 CEC Cherry second_postpool 1.61441010 0.02863731 0.35854062
## 8 CR1 Cranberry first_prepool -0.41891064 0.06108829 0.26983737
## 9 CR2 Cranberry first_prepool -0.44425078 0.02665804 -0.10684877
## 10 CR3 Cranberry first_prepool -0.08878235 0.03934015 0.05809882
## 11 CR4 Cranberry first_prepool -0.31794722 0.05591827 0.35156311
## 12 CR5 Cranberry first_prepool -0.44134580 0.07287785 0.18840336
## 13 CRA Cranberry second_postpool 0.49072243 0.02497960 -0.14885350
## 14 CRB Cranberry second_postpool 0.60455161 0.02677848 0.29719689
## 15 CRC Cranberry second_postpool 0.64866188 0.02340626 0.02703717
## 16 CRD Cranberry second_postpool 0.55460135 0.02652983 -0.15627744
## 17 CRE Cranberry second_postpool 0.75471935 0.02312381 0.72513486
## 18 FR1 Strawberry first_prepool 0.39149042 0.03843131 0.33051044
## 19 FR2 Strawberry first_prepool -0.24506580 0.06386743 0.09022737
## 20 FR3 Strawberry first_prepool 0.08495270 0.03464317 0.13335674
## 21 FR4 Strawberry first_prepool 0.13293173 0.03338454 0.58063316
## 22 FR5 Strawberry first_prepool -0.41701639 0.03195222 0.18650420
## 23 FRA Strawberry second_postpool 0.86504109 0.03421319 0.15709032
## 24 FRB Strawberry second_postpool 0.96698510 0.03291146 0.56023100
## 25 FRC Strawberry second_postpool 1.05024613 0.03374030 0.36640738
## sd_logchange Line_type mateff
## 1 0.25731181 dashed 1.81035697
## 2 0.57349396 solid 4.89102546
## 3 0.11827433 dashed 0.60365285
## 4 0.24311831 dashed 0.67875364
## 5 0.15742840 dashed 0.90608970
## 6 0.10098643 dashed 0.83920293
## 7 0.11692997 dashed 0.77791231
## 8 0.21318308 dashed 1.64414066
## 9 0.46435229 solid 0.75948546
## 10 0.13676999 dashed 1.65439607
## 11 0.12714934 dashed 2.10572789
## 12 0.14497197 dashed 1.42688378
## 13 0.11935520 dashed 1.30333542
## 14 0.15311973 dashed 0.50840113
## 15 0.11563812 dashed 0.95831855
## 16 0.11566948 dashed 1.28178337
## 17 0.12754300 dashed 0.03919933
## 18 0.08222221 dashed 0.15576367
## 19 0.17892718 dashed 1.36817612
## 20 0.16884574 dashed -0.56977628
## 21 0.13295026 dashed -3.36790492
## 22 0.25058617 dashed 1.44723471
## 23 0.14115599 dashed 0.81840132
## 24 0.12842826 dashed 0.42064154
## 25 0.12011698 dashed 0.65112237#Plot
PLOT_Maternaleffects<-ggplot(data=data_compare_longitudinal_pheno,
aes(y=mateff,x=logchange_long, color=Fruit_s, shape=Fruit_s, fill=interaction(Phase,Fruit_s))) +
geom_point(size=3, stroke=1.3) +
scale_fill_manual(name = "Phase",
values = c("#ffffff","#BC3C6D", "#ffffff","#FDB424", "#ffffff", "#3FAA96"),
breaks = c("first_prepool.Cherry", "second_postpool.Cherry"),
labels = c("Phase 1","Phase 3")) +
scale_color_manual(values=c("#BC3C6D", "#FDB424", "#3FAA96")) +
scale_shape_manual(values=c(21, 22, 24)) +
xlab("Fitness difference between average fitness\nacross generations 2 to 5 (13 to 26) and generation 2") +
ylab("Proportion of adaptation\ndue to evolution of maternal effects") +
labs(color="Selection fruit", shape="Selection fruit") +
guides(fill = guide_legend(override.aes = list(fill = c("white","black"), shape=21))) +
theme_LO_sober
PLOT_Maternaleffects 
cowplot::save_plot(file=here::here("figures", "SUPMAT_Maternal_effects.pdf"),
PLOT_Maternaleffects, base_height = 20/cm(1), base_width = 15/cm(1), dpi = 1200)4 SUP: Adaptation 8fruits
4.1 Longitudinal data
data_sum_allfruits <- loadlongitudinaldata_allfruits(dataset = "DATA_Adults_G1G29.csv", rm_generation_max = 5)
data_sum_allfruits_AN<-data_sum_allfruits[data_sum_allfruits$Generation!=1,]4.2 Analysis
#Models
mod_all_interaction <- lme4::lmer(fitness ~ Fruit_s*Generation + (1|Generation) + (1|Generation:Fruit_s),
weights = N, data = data_sum_allfruits_AN)
mod_all_fruit_generation<- lme4::lmer(fitness ~ Fruit_s+Generation + (1|Generation) + (1|Generation:Fruit_s) ,
weights = N, data = data_sum_allfruits_AN)
mod_all_fruit<- lme4::lmer(fitness ~ Fruit_s + (1|Generation) + (1|Generation:Fruit_s) ,
weights = N, data = data_sum_allfruits_AN)
mod_all_null<- lme4::lmer(fitness ~ 1 + (1|Generation) + (1|Generation:Fruit_s),
weights = N, data = data_sum_allfruits_AN)
MuMIn::model.sel(mod_all_interaction,mod_all_fruit_generation,mod_all_fruit,mod_all_null)## Model selection table
## (Int) Frt_s Gnr Frt_s:Gnr family df
## mod_all_fruit -2.4510 + gaussian(identity) 11
## mod_all_fruit_generation -2.4100 + -0.02122 gaussian(identity) 12
## mod_all_interaction -1.2580 + -0.56070 + gaussian(identity) 19
## mod_all_null -0.9595 gaussian(identity) 4
## logLik AICc delta weight
## mod_all_fruit -98.116 221.0 0.00 0.942
## mod_all_fruit_generation -99.726 226.8 5.76 0.053
## mod_all_interaction -92.293 231.3 10.31 0.005
## mod_all_null -115.017 238.4 17.42 0.000
## Models ranked by AICc(x)
## Random terms (all models):
## '1 | Generation', '1 | Generation:Fruit_s'#Posthoc
emmeans::emmeans(mod_all_fruit, list(pairwise ~ Fruit_s), adjust = "tukey") ### $`emmeans of Fruit_s`
## Fruit_s emmean SE df lower.CL upper.CL
## Blackcurrant -2.451 0.286 356 -3.237 -1.6654
## Cherry -0.528 0.168 423 -0.987 -0.0679
## Cranberry -0.483 0.155 487 -0.907 -0.0595
## Fig -0.963 0.196 347 -1.501 -0.4253
## Grape -2.437 0.357 272 -3.419 -1.4553
## Rosehips -2.076 0.285 367 -2.859 -1.2934
## Strawberry -0.424 0.160 426 -0.864 0.0148
## Tomato -1.505 0.338 149 -2.441 -0.5692
##
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
## Conf-level adjustment: sidak method for 8 estimates
##
## $`pairwise differences of Fruit_s`
## contrast estimate SE df t.ratio p.value
## Blackcurrant - Cherry -1.9237 0.328 348 -5.857 <.0001
## Blackcurrant - Cranberry -1.9681 0.323 361 -6.102 <.0001
## Blackcurrant - Fig -1.4882 0.342 321 -4.346 0.0005
## Blackcurrant - Grape -0.0141 0.448 261 -0.032 1.0000
## Blackcurrant - Rosehips -0.3750 0.397 313 -0.945 0.9813
## Blackcurrant - Strawberry -2.0267 0.325 350 -6.235 <.0001
## Blackcurrant - Tomato -0.9460 0.433 181 -2.185 0.3658
## Cherry - Cranberry -0.0444 0.226 381 -0.196 1.0000
## Cherry - Fig 0.4355 0.255 329 1.709 0.6813
## Cherry - Grape 1.9096 0.391 282 4.884 <.0001
## Cherry - Rosehips 1.5487 0.327 349 4.733 0.0001
## Cherry - Strawberry -0.1030 0.229 358 -0.449 0.9998
## Cherry - Tomato 0.9777 0.373 167 2.624 0.1547
## Cranberry - Fig 0.4799 0.247 349 1.943 0.5222
## Cranberry - Grape 1.9539 0.386 287 5.060 <.0001
## Cranberry - Rosehips 1.5931 0.321 364 4.957 <.0001
## Cranberry - Strawberry -0.0587 0.221 383 -0.266 1.0000
## Cranberry - Tomato 1.0220 0.368 169 2.780 0.1072
## Fig - Grape 1.4741 0.403 271 3.662 0.0072
## Fig - Rosehips 1.1132 0.341 319 3.263 0.0265
## Fig - Strawberry -0.5386 0.250 332 -2.152 0.3841
## Fig - Tomato 0.5421 0.384 164 1.411 0.8509
## Grape - Rosehips -0.3609 0.448 267 -0.806 0.9927
## Grape - Strawberry -2.0126 0.388 283 -5.184 <.0001
## Grape - Tomato -0.9319 0.480 175 -1.943 0.5235
## Rosehips - Strawberry -1.6518 0.324 353 -5.101 <.0001
## Rosehips - Tomato -0.5711 0.432 180 -1.322 0.8896
## Strawberry - Tomato 1.0807 0.370 167 2.923 0.0745
##
## Degrees-of-freedom method: kenward-roger
## P value adjustment: tukey method for comparing a family of 8 estimatesmod_all_fruit<- lme4::lmer(fitness ~ Fruit_s -1+ (1|Generation) + (1|Generation:Fruit_s) ,
weights = N, data = data_sum_allfruits)
summary(mod_all_fruit)## Linear mixed model fit by REML ['lmerMod']
## Formula: fitness ~ Fruit_s - 1 + (1 | Generation) + (1 | Generation:Fruit_s)
## Data: data_sum_allfruits
## Weights: N
##
## REML criterion at convergence: 258.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7627 -0.6555 0.1107 0.5521 2.8168
##
## Random effects:
## Groups Name Variance Std.Dev.
## Generation:Fruit_s (Intercept) 0.1504 0.3879
## Generation (Intercept) 0.1323 0.3637
## Residual 2.1944 1.4813
## Number of obs: 147, groups: Generation:Fruit_s, 33; Generation, 5
##
## Fixed effects:
## Estimate Std. Error t value
## Fruit_sBlackcurrant -1.7039 0.3351 -5.084
## Fruit_sCherry -0.3605 0.2572 -1.402
## Fruit_sCranberry -0.3561 0.2522 -1.412
## Fruit_sFig -0.8134 0.2799 -2.906
## Fruit_sGrape -1.9430 0.3575 -5.435
## Fruit_sRosehips -1.5403 0.3276 -4.701
## Fruit_sStrawberry -0.2591 0.2540 -1.020
## Fruit_sTomato -1.9056 0.3355 -5.680
##
## Correlation of Fixed Effects:
## Frt_sB Frt_sCh Frt_sCr Frt_sF Frt_sG Frt_sR Frt_sS
## Frut_sChrry 0.318
## Frt_sCrnbrr 0.320 0.409
## Fruit_sFig 0.316 0.375 0.382
## Fruit_sGrap 0.285 0.299 0.300 0.297
## Frut_sRshps 0.298 0.324 0.327 0.322 0.285
## Frt_sStrwbr 0.319 0.407 0.414 0.378 0.299 0.325
## Fruit_sTomt 0.288 0.316 0.319 0.314 0.275 0.290 0.3184.3 Plot
pd <- position_dodge(0.3) # move them .05 to the left and right
PLOT_ALLFRUITS<-ggplot(data=data_sum_allfruits,
aes(x=factor(Generation),group=Line, y=fitness, colour=Fruit_s)) +
facet_wrap(~ Fruit_s, nrow = 8) +
geom_errorbar(aes(ymin=fitness-1.96*se_fitness, ymax=fitness+1.96*se_fitness),
width=.1,position=pd,size=0.2,color="black") +
geom_line(size=0.3,position=pd) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylab("Fitness") +
xlab("Generation") +
theme_LO_sober
PLOT_ALLFRUITS
#Extract slope and intercept
dat_8fruits <- expand.grid(Generation=as.numeric(levels(as.factor(data_sum_allfruits_AN$Generation))),
Fruit_s=unique(data_sum_allfruits_AN$Fruit_s))
dat_8fruits$fitness_predicted <- predict(mod_all_fruit, newdata = dat_8fruits,
re.form= NA, type = "response")
PLOT_CHERRY <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Cherry",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Cherry",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylim(-5.65, 1.25) +
ylab("Fitness") +
xlab("Generation") +
xlim("1","2","3","4","5") +
scale_color_manual(values = c("#BC3C6D")) +
ggtitle("Cherry") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#BC3C6D"))
PLOT_CHERRY
PLOT_CRANBERRY <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Cranberry",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Cranberry",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylim(-5.65, 1.25) +
ylab("Fitness") +
xlim("1","2","3","4","5") +
xlab("Generation") +
scale_color_manual(values = c("#FDB424")) +
ggtitle("Cranberry") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#FDB424"))
PLOT_CRANBERRY
PLOT_STRAWBERRY <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Strawberry",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Strawberry",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c("#3FAA96")) +
ggtitle("Strawberry") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#3FAA96"))
PLOT_STRAWBERRY
PLOT_GRAPE <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Grape",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Grape",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c("#7B7554")) +
ggtitle("Grape") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#7B7554"))
PLOT_GRAPE
PLOT_ROSEHIPS <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Rosehips",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Rosehips",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white",stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c("#8ACDEA")) +
ggtitle("Rose hips") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#8ACDEA"))
PLOT_ROSEHIPS
PLOT_TOMATO <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Tomato",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Tomato",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white", stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c( "#ED6F47")) +
ggtitle("Tomato") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#ED6F47"))
PLOT_TOMATO
PLOT_FIG <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Fig",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Fig",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white", stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c("#93194F")) +
ggtitle("Fig") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#93194F"))
PLOT_FIG
PLOT_BLACKCURRANT <- ggplot(data = data_sum_allfruits[data_sum_allfruits$Fruit_s == "Blackcurrant",],
aes(x = factor(Generation), group = Line, y = fitness, colour =Fruit_s)) +
geom_errorbar(aes(ymin = fitness-1.96*se_fitness, ymax = fitness + 1.96*se_fitness),
width=.1,position = pd, size = 0.4,color = "black") +
geom_line(size = 0.5,position = pd) +
# geom_line(data = dat_8fruits[dat_8fruits$Fruit_s == "Blackcurrant",],
# aes(x = factor(Generation), y = fitness_predicted, colour = "black", group=NA), size = 0.7) +
geom_point(size = 2, position = pd, shape = 21, fill = "white", stroke = 1.2) +
ylim(-5.65, 1.25) +
xlim("1","2","3","4","5") +
ylab("Fitness") +
xlab("Generation") +
scale_color_manual(values = c( "#203753")) +
ggtitle("Blackcurrant") +
theme_LO_adaptation + theme(plot.title = element_text(color = "#203753"))
PLOT_BLACKCURRANT
DYNAMIC_EIGHT_JOIN <- cowplot::ggdraw() +
cowplot::draw_plot(PLOT_STRAWBERRY, x = 0, y = 0, width = 0.5, height = 0.28) +
cowplot::draw_plot(PLOT_GRAPE+ theme(axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0, y = 0.28, width = 0.5, height = 0.23) +
cowplot::draw_plot(PLOT_CRANBERRY+ theme(axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0, y = 0.53, width = 0.5, height = 0.23) +
cowplot::draw_plot(PLOT_BLACKCURRANT+ theme(axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0, y = 0.77, width = 0.5, height = 0.23) +
cowplot::draw_plot(PLOT_TOMATO+ theme(axis.title.y = element_blank()),
x = 0.5, y = 0, width = 0.5, height = 0.28) +
cowplot::draw_plot(PLOT_ROSEHIPS+ theme(axis.title.y = element_blank(),
axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0.5, y = 0.28, width = 0.5, height = 0.23) +
cowplot::draw_plot(PLOT_FIG+ theme(axis.title.y = element_blank(),
axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0.5, y = 0.53, width = 0.5, height = 0.23) +
cowplot::draw_plot(PLOT_CHERRY + theme(axis.title.y = element_blank(),
axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.line.x = element_blank()),
x = 0.5, y = 0.77, width = 0.5, height = 0.23) +
cowplot::draw_plot_label(c("A", "C", "E", "G", "B", "D", "F", "H"),
x = c(0, 0, 0, 0, 0.5, 0.5, 0.5, 0.5),
y = c(1, 0.75, 0.50, 0.27, 1, 0.75, 0.5, 0.27),
hjust = c(-0.5, -0.5, -0.5,-0.5, -0.5, -0.5, -0.5, -0.5),
vjust = c(1.5, 1.5, 1.5,1.5, 1.5, 1.5,1.5, 1.5),
size = 12)
DYNAMIC_EIGHT_JOIN
cowplot::save_plot(file =here::here("figures", "SUPMAT_Fitness8Fruits.pdf"), DYNAMIC_EIGHT_JOIN, base_height = 17/cm(1), base_width = 15/cm(1), dpi = 1200)